A few years ago, the cloud had many business leaders abuzz with the promise of lower costs and better performance. The cloud was touted as a cureall. Today, most businesses have adopted some cloud infrastructure and discovered that, like anything else, it’s not a magical solution. While it has many practical applications and can be transformative for businesses when applied correctly, the cloud won’t automatically or unilaterally fix the problems in any business.
The same is true for the latest crop of cloud-adjacent technologies – still, though, the hype around them mirrors what we saw around the cloud itself a few years ago. Here’s a quick overview of the buzz (and substance) behind three of the most talked-about developments.
Blockchain: mostly buzz
Blockchain is sometimes described as a solution in need of a problem. For now, that’s largely true.
While it’s a fascinating technology that will no doubt find practical applications, it’s just not practical in most business use cases. This lack of practicality for most businesses is because the blockchain is both immutable and distributed. Any business that hopes to use blockchain must be okay with sharing control of its data with third parties – which is a non-starter in many industries.
Still, blockchain is an interesting technology to keep an eye on, and may prove effective for certain applications where public data access is key. One really interesting pilot program currently in place is using blockchain to track medical licenses in Illinois.
AI & Machine Learning: useful but limited
The premise of AI and machine learning is undeniably cool: your computers can figure stuff out for you, making you better at what you do.
Today, more and more organizations are benefitting from AI included in cloud-based software or service providers (including Google, AWS, and Azure). Take Gmail, for example: ever noticed how the subscription emails you never open eventually migrate to your promotions or spam folder? That’s AI in action.
Even a few years on, AI use cases haven’t changed much: a Deloitte study out this year shows that 58 percent of “cognitive aware” companies are using AI primarily in the context of prepackaged software integrations.
The frustrating reason adoption isn’t likely to evolve beyond that in the near future is a skills gap: 37 percent of leaders said finding AI developers was a hurdle and 27 percent said the gap was a “major” or “extreme” hurdle.
Other frustrations leaders reported:
- Assessing data quality, cleaning data, and training AI (38 percent).
- Integrating AI with existing systems (38 percent).
Certain AI functionalities are ready to go, but leaders shouldn’t expect these to offer an edge; because they’re built to solve common business problems, they’re likely to become table stakes rather than a differentiator.
For machine learning specifically, the ROI problems from the cloud apply: while the tech is there to make ML possible and profitable, it requires an upfront investment to work. That means not only hiring the right talent but also knowing where and how data must be stored to get the kinds of findings that can make a material difference to a company’s bottom line. In other words: if you’re trying ML, go in with a well defined plan.
Internet of Things: useful
IoT use in supply chains and for system management and administration is growing rapidly, but, as Gartner notes, rarely as part of “a complete end-to-end process.” It’s in that end-to-end configuration, of course, that IoT will be most useful to businesses.
IoT in the supply chain is offering a lot of value, but IoT’s full potential and its ROI are still ahead of us.
